303 Commits

Author SHA1 Message Date
Gregory Nutt
cbdd590c82 Revert "include/nuttx/sched.h: Add storage for a previous signal mask. arch/: In all syscall implementations, block all signals before dispatching a system call; resotre signal mask when the system call returns."
Using the sigprocmask() for this purpose has too many side-effects.

This reverts commit 344f7bc9f61d139b932bf6c6a5001e7a8e54409d.
2019-11-28 11:57:54 -06:00
Gregory Nutt
344f7bc9f6 include/nuttx/sched.h: Add storage for a previous signal mask. arch/: In all syscall implemenations, block all signals before dispatching a system call; resotre signal mask when the system call returnes. 2019-11-28 10:51:29 -06:00
Nathan Hartman
68d43d3679 Fix typos. 2019-08-04 14:50:28 -06:00
Gregory Nutt
abf6965c24 Squashed commit of the following:
libs/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    syscall/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    wireless/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    Documentation/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    include/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    drivers/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    sched/:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    configs:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    arch/xtensa:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    arch/z80:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    arch/x86:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    arch/renesas and arch/risc-v:  Remove references to CONFIG_DISABLE_SIGNALS.  Signals can no longer be disabled.
    arch/or1k:  Remove all references to CONFIG_DISABLE_SIGNALS.  Signals are always enabled.
    arch/misoc:  Remove all references to CONFIG_DISABLE_SIGNALS.  Signals are always enabled.
    arch/mips:  Remove all references to CONFIG_DISABLE_SIGNALS.  Signals are always enabled.
    arch/avr:  Remove all references to CONFIG_DISABLE_SIGNALS.  Signals are always enabled.
    arch/arm:  Remove all references to CONFIG_DISABLE_SIGNALS.  Signals are always enabled.
2019-04-29 14:52:05 -06:00
Gregory Nutt
c1beda8a34 ARMv7-A/R: Add barriers.h header file for consistency with ARMv7-M. 2019-03-19 13:20:23 -06:00
Xiang Xiao
64252a298f arch/: Unify the cache interface for all architectures 2019-03-19 10:37:13 -06:00
Gregory Nutt
a2e62f557d Squashed commit of the following:
sched/init/nx_bringup.c:  Fix a naming collision.
    sched/init:  Rename os_start() to nx_start()
    sched/init:  Rename os_smp* to nx_smp*
    sched/init:  Rename os_bringup to nx_bringup
    sched/init:  rename all internal static functions to begin with nx_ vs os_
2019-02-04 16:20:35 -06:00
Gregory Nutt
bb623d1e04 This commit renames all internal OS functions defined under sched/task so that they begin with the prefix. For example, nxtask_exit() vs. task_exit().
Squashed commit of the following:

    Trivial, cosmetic
    sched/, arch/, and include:  Rename task_vforkstart() as nxtask_vforkstart()
    sched/, arch/, and include:  Rename task_vforkabort() as nxtask_vforkabort()
    sched/, arch/, and include:  Rename task_vforksetup() as nxtask_vfork_setup()
    sched/:  Rename notify_cancellation() as nxnotify_cancellation()
    sched/:  Rename task_recover() to nxtask_recover()
    sched/task, sched/pthread/, Documentation/:  Rename task_argsetup() and task_terminate() to nxtask_argsetup() and nxtask_terminate(), respectively.
    sched/task:  Rename task_schedsetup() to nxtask_schedsetup()
    sched/ (plus some binfmt/, include/, and arch/):  Rename task_start() and task_starthook() to nxtask_start() and nxtask_starthook().
    arch/ and sched/:  Rename task_exit() and task_exithook() to nxtask_exit() and nxtask_exithook(), respectively.
    sched/task:  Rename all internal, static, functions to begin with the nx prefix.
2019-02-04 13:42:51 -06:00
Gregory Nutt
a77c073797 arch/: Fix an interlock that was broken by commit 641a98a434b5819eba15b6ad494f0e3a8fa33978 in all implementations of up_sigdeliver. 2019-02-03 17:14:32 -06:00
Gregory Nutt
641a98a434 arch/: The saved return register state is available on the user stack. Thic commit reorders some logic so that certain, critical registers are preserved in the TCB. This does not make the logic 100% secure, but does prevent some obvious things. sched/signal/sig_delivery: Add a flag to the TCB to indicate that we are in a signal handler. Use this flag to assure that a there a never attempts to nest signal handling operations on a thread. This was guaranteed before but when locking of pre-emption during signal delivering was eliminated in a previous commit, there was a remote possibility of an attempt to do nested signal handling. This flag assures that there is only one signal handled at a time. 2019-02-03 15:29:47 -06:00
Xiang Xiao
763ba51b78 arch/xxx/src/xxx/up_assert.c: Dump CPU0 IDLE stack only when PID equals 0. IDLE threads of other CPUs do not need this special check 2019-01-26 10:59:23 -06:00
Gregory Nutt
db24306435 arch/assertion logic: Fix additional places where the test for an IDLE task is incorrect. It is not invalid in all configurations to check for PID==0. However, the logic fixed in these places lackes sufficient intelligence to find the right stack for the CPU IDLE thread and could still show the wrong stack. 2018-12-18 17:45:46 -06:00
Gregory Nutt
ec9265aa95 arch assertions: Correct duplicated logic from commit dbf01d12b7321d1b48b68d3b69842a853b049376. Checking for PID == zero is not a valid way to test for the IDLE task in all configurations! This is only true in the single CPU configuration. In multiple CPU configurations, there will be a separate IDLE task for each CPU with a different PID. 2018-12-18 17:35:42 -06:00
Xiang Xiao
dbf01d12b7 Assertions: Identify the running task correctly when dumping task state information. It takes time to switch to the target task after g_readytorun has been modified. If panic/assert happen during this period, the dump will contain the incorrect and confusing information due to the difference between the real running task and the return value of this_task(). This change resolve this problem by adding g_running_task to track the real running task through the context switch. 2018-11-15 07:11:51 -06:00
Xiang Xiao
543f4ed8ec arch/ all assertion functions: up_assert move the register dump to first make the more important info first 2018-11-11 12:53:59 -06:00
Xiang Xiao
dfe788be25 arch/ all assertion functinos: up_stackdump dump the full stack if stack overflow the stack info is very useful to find the backtrace 2018-11-11 12:52:36 -06:00
Xiang Xiao
e4106a3744 arch/ assertions files: up_registerdump capture the general register if not yet saved and up_saveusercontext is implemented, the register dump is very useful to find the cause of failure. 2018-11-11 12:50:50 -06:00
Gregory Nutt
2b3ec4172d arch/: Add 'BOARD_ASSERT_RESET_VALUE' in config/Kconfig and replace reboot status '0' to 'CONFIG_BOARD_ASSERT_RESET_VALUE'. 2018-11-10 14:06:46 -06:00
Gregory Nutt
b823fd83bd arch/arm/src/armv7-a: Replicate the same fix was previously commited for the armv7-r. 2018-09-20 21:40:29 -06:00
EunBong Song
0f18e8cc32 Merged in ebsong/nuttx (pull request #722)
Pull request for mtd/progmem refactoring.

* mtd/progmem: change up_progmem_npages to up_progmem_neraseblocks

    page is a unit for read/write operation.
    eraseblock is a unit for erase operation.
    up_progmem_npages is a little bit confusing because it returns number of
    erase blocks in flash memory. This patch changes up_progmem_npages to
    up_progmem_neraseblocks. There is no logical change.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* mtd/progmem: up_progmem_erasesize.

    Change argument name to be more readable.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* mtd/progmem: up_progmem_eraseblock

    Change up_progmem_erasepage to up_progmem_eraseblock.
    eraseblock is more readable name than erasepage.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* mtd/progmem: change up_progmem_eraseblock's return value.

    up_progmem_eraseblock erase a block. so it's better to return the erase block
    size than page size.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* arm/cortex-r : fix wrong cp15_flash_dcache.

    change mcrne to mcr for unconditional dcache.

    Signed-off-by: sungduk.cho <sungduk.cho@samsung.com>

Approved-by: GregoryN <gnutt@nuttx.org>
2018-09-21 03:18:38 +00:00
EunBong Song
d33b0640fe Merged in ebsong/nuttx (pull request #712)
Pull request for cortex-r4 codes

* arm/armv7-r: Add general interrupt controller.

    This is based on armv7-a gic controller code.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* arm/armv7-r: add invalidate dcache in arm_head.S

    Adding invalidate dcache as a comment in arm_head.S.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* arm/armv7-r: Fix some wrong configuration of program status register.

    PSR_E_BIT bit should be set for big endian system.
    PSR_A_BIT bis is set automatically as arm cortex-r4 reference manual 3.7.4.
    So we don't need to set this bit.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* arm/armv7-r: Fix some wrong MPU register definition.

    Change MPU_RBAR_ADDR_MASK and MPU_RACR_TEX_SHIFT mask as
    arm cortex-r4 reference manual.

    Region Base Address Register 0-4 bits are reserved.
    MPU Region Access control register type 3-5 bits.

    Signed-off-by: EunBong Song <eunb.song@samsung.com>

* driver/mtd: fix compilation error.

    This commit fixes below compilation errors.

    CC:  mtd/smart.c
    mtd/smart.c:182:22: error: 'gWearBitToLevelMap4' defined but not used [-Werror=unused-const-variable=]
     static const uint8_t gWearBitToLevelMap4[] =
                          ^~~~~~~~~~~~~~~~~~~
    mtd/smart.c:170:22: error: 'gWearLevelToBitMap4' defined but not used [-Werror=unused-const-variable=]
     static const uint8_t gWearLevelToBitMap4[] =
                          ^~~~~~~~~~~~~~~~~~~
    cc1: all warnings being treated as errors
    make[1]: *** [smart.o] Error 1

    Signed-off-by: Junyeon LEE <junyeon2.lee@samsung.com>

Approved-by: GregoryN <gnutt@nuttx.org>
2018-08-29 01:50:41 +00:00
Xiang Xiao
7a9309370f arm syscalls: svcall/sycall logic needs to get the ucontext argument from R4 instead of stack since all syscall parameters pass from registers in syscall.h 2018-08-24 07:11:18 -06:00
Xiang Xiao
e1202d2ed3 Replace all ASSERT with DEBUGASSERT to save the code space 2018-08-24 06:58:30 -06:00
Xiang Xiao
467d2a58ea Replace non critical PANIC with DEBUGPANIC to save the code space 2018-08-24 06:21:15 -06:00
Xiang Xiao
a465b6f0d4 configs/: Change CONFIG_BOARD_RESET_ON_CRASH to CONFIG_BOARD_RESET_ON_ASSERT, arch/: Implement call to board_reset() if in all implementations of up_assert() when CONFIG_BOARD_RESET_ON_ASSERT=y. 2018-08-22 17:04:39 -06:00
Gregory Nutt
a94e3284b3 syslog: Enable is partial, crippled version of syslog_flush(); arch/: Call syslog_flush() from assertion handling logic. 2018-06-07 16:29:16 -06:00
Gregory Nutt
977d41d519 Based on a change recommended by Mark Shulte:
Signal handlers maybe run with interrupts enabled or disabled, depending on how the task the received the signal was blocked. (i.e.: If sem_wait() is called, then we disable interrupts, then block the currently running task). This could be dangerous, because user code would be running with interrupts disabled.

This change forces interrupts to be enabled in up_sigdeliver() before executing the signal handler calling up_irq_enable() explicitly.  This is safe because, when we return to normal execution, interrupts will be restored to their previous state when the signal handler returns.
2018-06-06 09:54:30 -06:00
Gregory Nutt
50ca22ae78 Trivial, cosmetic update to last review. 2018-04-18 10:14:06 -06:00
Gregory Nutt
1efd591cd8 arch/arm/include/tms570, arm/src/armv7-r, and arm/src/tms570: Adds support for the TMS570LS3137ZWT and corrects seversl ARMv7-R and TMS570 issues 2018-04-18 08:58:36 -06:00
Gregory Nutt
de6b13b3ab Per OpenGroup.org, syslog -- and, hence, nonstandard vsyslog, and debug wrappers -- does not return a value. Rename _vsyslog to nx_vsyslog. Use internal nx_vsyslog in the few cases where a return value is required. 2018-03-04 08:07:07 -06:00
Gregory Nutt
5c5d19f1c8 Some ommissions from previous commit. 2018-02-04 12:36:51 -06:00
Gregory Nutt
1e59d9dd14 armv7-a, armv7-r, armv7-m: Add atomic read-add-write and read-subtract-write functions. 2018-02-04 12:22:03 -06:00
Gregory Nutt
1567b82429 Make sure that labeling is used consistently in all function headers (part 2). 2018-02-01 12:03:55 -06:00
Gregory Nutt
7cf88d7dbd Make sure that labeling is used consistently in all function headers. 2018-02-01 10:00:02 -06:00
Gregory Nutt
6e4918c557 Remove CONFIG_ARM_TOOLCHAIN_GNU; replace with CONFIG_ARCH_TOOLCHAIN_GNU 2017-05-13 13:28:15 -06:00
Manohara HK
b154531838 I found an issue inside the cp15_coherent_dcache function in file, arch/arm/src/armv7-r/cp15_coherent_dcache.S.
The "mcr CP15_BPIALLIS(r0)" instruction is used for invalidating entire branch predictor.  But the problem is, since this is the generic code and can be called on any armv7-r architecture based CPU's.  It is a problem, if this instruction is called in uni processor configuration. Because, BPIALLIS (c7, 0, c1, 6) instruction is only added as part of the "Multiprocessing Extensions" (As per ARM® Architecture Reference Manual /ARMv7-A and ARMv7-R edition)

So in my opinion, this instruction should be under SMP configuration. In non-SMP configuration this instruction could become undefined.
2017-02-13 06:33:15 -06:00
Gregory Nutt
edd9186540 ELF: Move ARMv7-A and ARMv7-R versions of ELF relocation logic to libc/machine 2017-01-21 14:40:26 -06:00
Gregory Nutt
be5ba90d4f Move optimized ARM memcpy functions from arch/arm/src/ to libc/machine/. This is necessary for the PROTECTED and KERNEL build modes. Otherwise, memcpy() will be built in to kernel space and not accessible to applications. 2017-01-20 10:53:46 -06:00
Gregory Nutt
13d00344c9 Add configuration to prevent selection of Windows native toolchains when using Ubuntu under Windows 10 2017-01-02 07:16:47 -06:00
Gregory Nutt
3a0413c048 Back out most of 34be3e7c3c5885dd657712f92965f52214105d78 and update README again. Windows native tools cannot be used with Ubuntu under Windows 10 now. For Cygwin, that support depends on the 'cygpath -w' tool to convert POSIX paths to Windows paths. There is no corresponding tool for Ubuntu under Windows 10. 2017-01-01 16:29:03 -06:00
Gregory Nutt
34be3e7c3c Add configuration support for builds with Ubuntu under Windows 10 2017-01-01 15:34:23 -06:00
Gregory Nutt
dc79e35d65 For Cortex-A9, should also set ACTLR.FW in SMP mode to enble TLB and cache broadcasts. Does not fix SMP cache problem. 2016-12-07 09:06:41 -06:00
Gregory Nutt
89c3c20052 Merge remote-tracking branch 'origin/master' into misoc 2016-11-06 11:09:05 -06:00
Gregory Nutt
c1a687a4e5 Trivial changes from review of last PR 2016-11-06 08:11:01 -06:00
Heesub Shin
bda7d9ee4d ARMv7-R: fix to restore the Thumb flag in CPSR
Thumb flag in CPSR is not restored back when the context switch occurs
while executing thumb instruction.

Reported-by: Eunbong Song <eunb.song@samsung.com>
Signed-off-by: Byoungtae Cho <bt.cho@samsung.com>
Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 21:48:02 +09:00
Heesub Shin
343243c7c0 ARMv7-R: fix CPSR corruption after exception handling
A sporadic hang with consequent crash is observed when booting:

    arm_prefetchabort: Prefetch abort. PC: 04d34a00 IFAR: 04d34a00 IFSR: 00000008
    up_assert: Assertion failed at file:armv7-r/arm_prefetchabort.c line: 87 task: init
    up_dumpstate: Current sp: 004c3df0
    up_dumpstate: Interrupt stack:
    up_dumpstate:   base: 004c05fc
    up_dumpstate:   size: 00000800
    up_dumpstate: User stack:
    up_dumpstate:   base: 004c3f58
    up_dumpstate:   size: 00000fec
    up_dumpstate: User Stack
    up_stackdump: 004c3de0: 004a0d14 004c3df0 004c3f58 004a0d20 00000057 004c2c58 09000000 004a42a4
    up_stackdump: 004c3e00: 00000003 004c3e10 004a0f1c 004bbcef 33c44b00 004a0f28 04d34a00 00000008
    up_stackdump: 004c3e20: 00000008 004a01bc 004bfd38 00000001 00007fff 00000001 34134a00 d83e4c00
    up_stackdump: 004c3e40: 09000000 00000000 33c44b00 d83e4c00 0c3f4c00 00000000 00000000 00000003
    up_stackdump: 004c3e60: 004c3e70 004a1494 04d34a00 200b0253 004c3ed8 004a5298 004c3ed8 6d00006d
    up_stackdump: 004c3e80: 0000006d 004bc3f4 00000009 004a4f64 00000009 b9e0784f 333f3ed0 69d4227d
    up_stackdump: 004c3ea0: d81f09bd 0f867344 5a7e2c12 8acefd34 5d00dc1b 004bc432 004c3f08 004c0e08
    up_stackdump: 004c3ec0: 00000000 00000000 00000000 00000000 00000000 004a4210 004a5258 004a5300
    up_stackdump: 004c3ee0: 00000000 00000001 ffffffff 004c3f80 000002b0 004a4234 00000007 004c3f08
    up_stackdump: 004c3f00: 004a58b4 004bc432 004bc3f4 004c3f80 000002b0 0000029c 00000000 0000029c
    up_stackdump: 004c3f20: 00000000 004a5900 0000ff01 00000000 00000000 004a61f4 00000000 004a5fa4
    up_stackdump: 004c3f40: 00000000 004a5f6c 00000000 004a2668 00000000 00000000 b7509f04 004c3f64
    up_registerdump: R0: 00000001 00007fff 00000001 34134a00 d83e4c00 09000000 00000000 33c44b00
    up_registerdump: R8: d83e4c00 0c3f4c00 00000000 00000000 00000003 004c3e70 004a1494 04d34a00
    up_registerdump: CPSR: 200b0253

It seems to be caused by the corrupted or wrong CPSR restored on return
from exception. NuttX restores the context using code like this:

    msr spsr, r1

GCC translates this to:

    msr spsr_fc, r1

As a result, not all SPSR fields are updated on exception return. This
should be:

    msr spsr_fsxc, r1

On some evaluation boards, spsr_svc may have totally invalid value at
power-on-reset. As it is not initialized at boot, the code above may
result in the corruption of cpsr and thus unexpected behavior.

Reported-by: Eunbong Song <eunb.song@samsung.com>
Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 20:48:09 +09:00
Heesub Shin
6bfc6b4d23 ARMv7-R: fix typo in mpu support
s/ARMV7M/ARMV7R/g

Reported-by: Eunbong Song <eunb.song@samsung.com>
Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 20:48:09 +09:00
Heesub Shin
003511d198 ARMv7-R: add cache handling functions
This commit adds functions for enabling and disabling d/i-caches which
were missing for ARMv7-R.

Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 20:48:01 +09:00
Heesub Shin
6a1a846011 ARMv7-R: add new Kconfig entries for d/i-cache
Unlike in ARMv7-A/M, Kconfig entries for data and instruction caches
are currently missing in ARMv7-R. This commit adds those missing Kconfig
entries. Actual implmenetation for those functions will be added in the
subsequent patches.

Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 16:07:46 +09:00
Heesub Shin
2b922fcdbd ARMv7-R: remove the redundant update on SCTLR
mpu_control() is invoking cp15_wrsctlr() around SCTLR update
redundantly.

Signed-off-by: Heesub Shin <heesub.shin@samsung.com>
2016-11-06 16:07:45 +09:00